The Dignity of Legislation

Maryland Law Review
Volume 54 | Issue 2 Article 12
The Dignity of Legislation

Jeremy Waldron
Follow this and additional works at: http://digitalcommons.law.umaryland.edu/mlr

Part of the Legislation Commons
Recommended Citation
Jeremy Waldron, The Dignity of Legislation , 54 Md. L. Rev. 633 (1995)
Available at: http://digitalcommons.law.umaryland.edu/mlr/vol54/iss2/12
This Casenotes and Comments is brought to you for free and open access by the Academic Journals at DigitalCommons@UM Carey Law. It has been
accepted for inclusion in Maryland Law Review by an authorized administrator of DigitalCommons@UM Carey Law. For more information, please
contact smccarty@law.umaryland.edu.
Essay
THE DIGNITY OF LEGISLATION
JEREMY WAMRON*
Civil Society being a State of Peace, amongst those who are

of it, from which the State of War is excluded by the Umpir-
age, which they have provided in their Legislative, for the
ending all Differences, that may arise amongst any of them,
'tis in their Legislative, that the Members of a Commonwealth
are united, and combined together into one coherent living
Body. This is the Soul that gives Form, Life, and Unity to the
Commonwealth: From hence the several Members [of soci-
ety] have their mutual Influence, Sympathy, and Connex-
ion.... For the Essence and Union of the Society consisting in
having one Will, the Legislative, when once established by
the Majority, has the declaring, and as it were keeping of that
Will.,
I doubt that anyone will say this about the 103rd Congress of the
United States: "This is the Soul that gives Form, Life, and Unity to the
Commonwealth."' And although I have been studying the Two Trea-
tises for more than twenty years, I was surprised to find that this was
John Locke's view. Indeed, I was taken aback when a student drew
this passage to my attention,3 because I had been presenting John
* Professor of Law, Jurisprudence and Social Policy Program, School of Law (Boalt
Hall), and Professor of Philosophy, University of California at Berkeley. B.A., 1974; LL.B.,
1978, University of Otago, New Zealand; M.A., 1981; D. Phil., 1986, University of Oxford.
This Essay is an expanded version of the Gerber Lecture given at the University of
Maryland School of Law on October 12, 1994. Earlier versions were presented at the Co-
lumbia Legal Theory Workshop, and at workshops at the University of Otago Law School
and the Research School of Social Sciences at the Australian National University. I am
grateful to all who offered suggestions and criticisms on those occasions and in subsequent
conversations and correspondence. Particular thanks to Jim Allen, Geoffrey Brennan,
Tom Campbell, Jules Coleman, Meir Dan-Cohen, George Fletcher, Don Gifford, Daniel
Goldberg, Kent Greenawalt, Bonnie Honig, Sanford Kadish, Martin Krygier, Rae Langton,
Charles Larmore, Michael Meyer, Stephen Perry, Richard Pierce, Nelson Polsby, Robert
Post, Eric Rakowski, Joseph Raz, Deborah Rhode, Dan Rodriguez, Alan Ryan, Carol
Sanger, Philip Selznick, Quentin Skinner, John Smillie, Peter Strauss, and Paul Thomas.
1. JoHN LOCKE, Two TREAnTsEs OF GovERmMEwsr 425 (Peter Laslett ed., Cambridge
Univ. Press 1970) (3d ed. 1698) (Second Treatise 212).
2. Id.
3. The student is now Professor Jacqueline Stevens, of the University of Michigan's
Political Science Department.
634 MARYLAND LAW REVIEW [VOL. 54:633
Locke as the great proponent of natural right and accordingly as a

believer in limited legislatures, the father of modem constitutional-
ism, perhaps even as a precursor of the American tradition of allowing
or encouraging the judicial review of legislation in the name of indi-
vidual rights.
I was not wrong: Locke is all those things. He did set quite strict4
limits to what he called "the extent of the Legislative Power."
Though perhaps we should not refer to him as a precursor of judicial
review. He was adamant that the Legislature ought to be "the
Supreme Power" in every society, at least so long as government ex-
ists,' and he maintained that when "the Legislative act contrary to [its]
trust," the proper remedy is revolution, not litigation.6
Still, the passage I quoted is a striking one, as is our neglect of it
in legal and political philosophy. For it is remarkable that, in our phi-
losophy of government, we spend much more time discussing the ap-
propriate limits on legislation-and the moral and philosophical basis
of those limits in the idea of individual rights-than we do discussing
the nature and character of the legislative power itself and its impor-
tance for a self-governing society. We are so confident that John
Locke shares our attitudes about constitutionalism and the separation
of powers, 7 that very few Locke scholars, very few political philoso-
phers of any complexion, have bothered to explore or consider or
even criticize the claim made in the passage I quoted-viz., that the
institution which makes our laws, and the institution which resolves
our ultimate differences in moral principle, ought to be one and the
same, and that the institution which combines these functions thereby
embodies our civic unity and our sense of mutual sympathy. "This"-
the Legislature-"is the Soul that gives Form, Life, and Unity to the
Commonwealth."8
Locke's hypothesis is a challenging one, but it is not the chal-

lenge I am going to take up in this Essay. Instead I want to draw atten-
tion to another feature of legislatures and legislation-a feature which
4. LocKE, supra note 1, at 373 (Second Treatise 134).

5. Id. at 385 (Second Treatise 150). "In all Cases, whilst the Government subsists,
the Legislative is the Supream Power." Id.
6. Id. at 385 (Second Treatise 149); see also id. at 430-32 (Second Treatise 221-
22).
7. See Samuel Freeman, ConstitutionalDemocracy and the Legitimacy ofJudicialReview, 9
LAw & PHIL. 327, 349 nn.21-22 (1990); ef.Jeremy Waldron, Freeman'sDefense ofJudicialRe-
view, 13 LAw & PHIL. 27, 33 n.8 (1994).
8. LocK, supra note 1, at 425 (Second Treatise 212).
1995] THE DIGNITY OF LEGISLATION
provides a sort of infrastructure or underpinning -for Locke's idea,

and which has also been systematically neglected, left beyond the pale
of discussion, in modern legal philosophy.
Philosophy proceeds by making familiar facts seem strange to us.
So consider this familiar fact: Everywhere in the modern world, legis-
latures are institutions which comprise hundreds of members, mem-
bers who take their decisions collectively, and deal with one another
formally as equals. 9
If we think about this fact at all, we take it as an obvious feature of
a legislature's democratic character, of its claim to be a representative
institution. We should remember, however, that democratic creden-
tials and theories of representation come in all shapes and sizes. A
single elective ruler might have democratic credentials in virtue of
having more popular support than any other candidate for his office.
In this regard, one person might be regarded as the representative of
the whole community: an elected President is an example. Alterna-
tively, an assembly of several hundred men and women might have
democratic credentials, not just in virtue of the fact that each of them
won an election, but because taken together their diversity represents,
along various dimensions, the diversity of the community at large.
Their democratic credentials consist partly in the fact that when they
talk to one another, different sections of society (different regions,
different interests, different ethnic groups) can be taken, through
their representation, to be talking to one another. In their plurality,
they represent the larger plurality of the community-as opposed to
the elective ruler who, in his or her singularity, represents at best the
hegemony of a single view or faction among the people.
Taken to an extreme, the second model of representativeness is
a pictorial one. As such, it has a respectable pedigree. John Adams
wrote that a representative legislature "should be an exact portrait, in
miniature, of the people at large."' ° James Wilson maintained, at the
Constitutional Convention, that legislative proceedings ought to be a
"faithful echo of the voices of the people,""1-of the voices plural, not
just of a unitary vox populi. On this conception, a representative as-
9. Cf the definition of "legislatures" in THE BLACKwELL ENCYCLOPEDIA OF POLITICAL

INSTITUTIONS 329 (Vernon Bogdanor ed., 1987) ("legislatures: Political institutions whose
members are formally equal to one another, whose authority derives from a claim that the
members are REPRESENTATIVES of the political community, and whose decisions are
collectively made according to complex procedures.").
10. Letter from John Adams to John Penn, in IV WORKS OF JOHN ADAMS 205 (Boston
1852-65), quoted in HANNA F. P1TKIN, THE CONCEPT OF REPRESENTATION 60 (1972).
11. PrrmN, supra note 10, at 61.
MARYLAND LAw REVwEw [VOL. 54:633
sembly ought to be a sort of microcosm, mapping to scale the features

and differences that the wider community exhibits:
as the map represents mountains and valleys, lakes and riv-
ers, forests and meadows, the legislative body, too, is to form
again a condensation of the component parts of the People,
as well as of the People as a whole, according to their actual
relationships.... The value of each part is determined by its
significance in the whole and for the whole.' 2
Obviously, there are limits on how far one can pursue this conception.
It is not at all clear, for example, if we desire representation of this
kind, that electoral competition is the best way of choosing represent-
atives. Maybe random selection or statistical sampling would be
preferable.' 3
Nevertheless, almost all modem legislatures claim their demo-
cratic credentials in virtue of their inherent plurality, and thus implic-
idy in terms of the second model of representativeness rather than the
first. No doubt, we should qualify this by observing that subordinate
legislation is often made by single individuals or by very small rule-
making agencies. This, however, should not distract us: such individ-
uals and agencies derive their authority from a sovereign legislature
that comprises hundreds of members. To state the point more care-
fully, we can say that everywhere sovereign legislatures are assemblies
rather than individuals. In almost every legal system, legislation bases
its final authority as law on the fact that it is the product of (or its
production has been authorized by) a large popular assembly.
This contrasts quite markedly with the other great institutions of
government. The highest court in each system characteristically com-
prises a very small number of judges and the executive is often just
one person or a small cabinet. Again, we should not be distracted by
the fact that large numbers of people are organized hierarchically
under these officials: inferior courts and their staffs; subordinate ex-
ecutive officers. What is remarkable about the legislature is that the
apex or, if you like, the highest rung in its hierarchy is occupied by
hundreds of persons, each of whom counts as the equal of each of the
others so far as their final law-making authority is concerned. In legis-
lation above all, particularly legislation at the highest level, we think it
important that authority be plural, and that people be represented by
people and not just by an individual person.
12. Id. at 62. This description is taken from the eighteenth century Swiss legal theorist
J.C. Bluntschli.
13. See id. at 73-75 and references cited therein.
I labor this point, to emphasize that the constitutional preference

for legislative assemblies survives, despite a clear consensus in the ca-
non of political philosophy that the size of a legislative body is an ob-
stacle, rather than an advantage, to rational law-making. Part of that
consensus is explained by a sense, originating in ancient prejudice but
surfacing also in the Enlightenment, that the larger the legislative as-
sembly the lower the average level of wisdom and knowledge among
the law-makers. The views of the Marquis de Condorcet are typical.
On the one hand, Condorcet proved arithmetically that majority-rule
makes a group more likely to give correct answers to some question
than the average member of the group; what is more, the greater the
size of the group, the more likely it is that the majority answer is right,
provided the average competence of the individual members of the
group (the chances of each coming up with the right answer to the
question before them) is greater than 0.5.14 On the other hand, Con-
dorcet also maintained that average individual competence tends in-
dependently to decline as group size increases (and then of course
the arithmetic of majority decision works in the other direction):
A very numerous assembly cannot be composed of very en-
lightened men. It is even probable that those comprising
this assembly will on many matters combine great ignorance
with many prejudices. Thus there will be a great number of
questions on which the probability of the truth of each voter
will be below 1/2. It follows that the more numerous the
14. Suppose there are three voters-V, W, and X-voting independently, each with a
0.6 chance of being right about the question they are addressing. When Vcasts her vote,
there is a 0.6 chance she is right and a 0.4 chance she is wrong. When Wcasts her vote,
there is a 0.6 x 0.6 = 0.36 chance that a majority comprising at least Wand Vwill be right, a
0.6 x 0.4 = 0.24 chance that Vwill be right and Wwrong, and a 0.4 x 0.6 0.24 chance that V
will be wrong and Wright. (There is also a 0.16 chance that they are both wrong.) Now
the third voter, X casts her vote. If Vgot it right and Wwrong, there is a 0.24 x 0.6 0.144
chance that a majority comprising only Vand Xwill be right. And if Vgot it wrong and W
right, there is the same chance (0.144) that a majority comprising only Wand Xwill be
right. The overall probability that a majority will be right then is 0.36 (VWX or VW) +
0.144 (VX) + 0.144 (WX) = 0.648, which is higher than the 0.6 individual competence we
began with.
For a sense of the difference that an increase in group size can make, consider that if
we add to the group two additional voters of the same individual competence (0.6), we get
a competence of 0.68256 for the five members deciding by a majority. To get a group
competence of higher than 0.9, we need only add an additional 36 members with individ-
ual competencies of 0.6. See Bernard Grofman & Scott Feld, Romseau's General Wilk A
CondorcetianPerspective, 82 AM. POL. So. Rzv. 567, 571 (1988).
MARYLAND ILAw REVIEW [VOL. 54:633
assembly, the more

15
it will be exposed to the risk of making
false decisions.
Even if the ignorance of the large group of legislators is not a
problem, there is still a concern, exhibited for example by James
Madison, about their susceptibility to passion and malign influence:
[T] he more multitudinous a representative assembly may be
rendered, the more it will partake of the infirmities incident
to collective meetings of the people. Ignorance will be the
dupe of cunning, and passion the slave of sophistry and dec-
lamation. The people can never err more than in supposing
that by multiplying their representatives beyond a certain
limit they strengthen the barrier against the government of a
few. Experience will forever admonish them that, on the
contrary, after securing a certain numberfor the purposes of safety,
of local information, and of diffusive sympathy with the whole soci-
ety, they will counteract their own views by every addition to
their representatives. The countenance of the government
may become more democratic, but the soul that animates it
will be more oligarchic.1 6
The apprehension about oligarchy reflects, in part, a concern
about the difficulty of large numbers of representatives coordinating
sufficiently to produce laws that are coherent. Rend Descartes
presented this as a problem of coordination over time, comparing
legal systems which have grown up over generations unfavorably with
those that were founded all at once according to a single master plan:
I thought to myself that the peoples who.., became civilized
only gradually, making their laws only insofar as the harm
done by crimes and quarrels forced them to do so, could not
be as well organized as those who, from the moment at
which they came together in association, observed the basic
laws of some wise legislator. . . . I believe that, if Sparta
greatly flourished in times past, it was not on account of the
excellence of each of its laws taken individually, seeing that
many were very strange and even contrary to good morals,
15. CONDORCET: SELECTED WITINGs 49 (Keith M. Baker ed., Bobbs-Merrill 1976)

(1785). See also Symposium, Democratic Theory and the Public Interest: Condorcet and Rousseau
Revisited, 83 AM. POL. Sci. REv. 1317, 1324 (1989).
16. THE FEDERALIST No. 58, at 351 (James Madison) (Isaac Kramnick ed., 1987). I am
grateful to Marshall Sanger for this reference. The whole discussion in THE FEDERALIST
Nos. 55-58 is worth reading in this connection.
19951 THE 'DIGNITY OF LEGISLATION 639
but because, having been invented by one man only, they all
tended towards the same end. 7
The most common concern, however, is about the coordination of
large numbers of members in a given legislative session. William
Blackstone described the task of extracting a "system from the discor-
dant opinions of more than five hundred counsellors" in a representa-
tive assembly as "Herculean."" And when Jean-Jacques Rousseau
asked, in The Social Contract, "How can a blind multitude, which often
does not know what it wills... carry out for itself so great and difficult
an enterprise as a system of legislation?" 9 it was partly this difficulty of
co-ordination that he had in mind. He addressed it with his image of
"the law-giver" a mythic figure distinguished, in this context, as much
by his singularity as by his "superior intelligence."" °
A hundred years later, we hear echoes of the same concerns, now
in English political theory. John Stuart Mill worried about the pros-
pects of coherent legislation emerging when bills are "voted clause by
clause in a miscellaneous assembly." 1 He argued in general that
"[n]o body of men, unless organized or under command, is fit for
action"22; and since legislative functions are as much matters of action
as executive functions, he concluded that "a numerous assembly is as
little fitted for the direct business of legislation as for that of adminis-
tration. " " Walter Bagehot wrote in similar fashion about the House
of Commons: "Here are 658 persons, collected from all parts of Eng-
land [sic], different in nature, different in interests, different in look
and language."2 4 How is something coherent supposed to emerge
from the babel of their cross-cutting proposals and counter-proposals?
There is a saying in England, Bagehot added, "'a big meeting never
17. RENE DESCARTES, DISCOURSE ON METHOD 36 (F.E. Sutcliffe trans., Penguin Books
1968) (1637), quoted-though not with approval-in MICHAEL WALZER, INTERPRETATION
AND SOCIAL CRITICISM 9-10 (1987).
18. 3 W. BLACKSTONE, COMMENTARIES 267, quoted in DAVID LIEBERMAN, THE PROVINCE OF
LEGISLATION DETERMINED: LEGAL THEORY IN EIGHTEENTH CENTURY BRITAIN 61 (1989).
19. JEAN-JACQUES ROUSSEAU, THE SOCIAL CONTRACT II. vi, at 37 (G.D.H. Cole trans.,
1950) (1762).
20. Id. at 39.
21. JOHN STUART MILL, CONSIDERATIONS ON REPRESENTATIVE GOVERNMENT, ch. V, at 77
(Currin V. Shields ed., Bobbs-Merrill 1958) (3d ed. 1865). Mill continued: "The incon-
gruity of such a mode of legislating would strike all minds, were it not that our laws are
already, as to form and construction, such a chaos, that the confusion and contradiction
seem incapable of being made greater by any addition to the mass." Id.
22. Id. at 71.
23. Id. at 76.
24. WALTER BAGEHOT, THE ENGLISH CONSTITUTION 138 (Garland Publishing 1978)
(1878).
640 MARYLAND LAW RE [VOL. 54:633
5w
does anything;' and yet we are governed by the House of Commons-

25
by 'a big meeting."
Nor have these concerns diminished in the twentieth century.
On the contrary, empirical political science and public choice theory
have raised our understanding of the difficulties of collective action to
a new level. Interest group theorists see legislation as, at best, bargain-
ing and deal-making between the representatives of more or less pow-
erful private interests. The understanding afforded by public choice
theory may be even less optimistic than that: a large body such as a
legislature, making its decisions by voting, may not have enough col-
lective rationality even to settle on a stable deal, let alone articulate
the public interest. What emerges from the chaos of the legislative
process may well be a matter of who controls the agenda or, even
more arbitrarily, of where in the cycle of voting a "decision" happens
26
to pop out.
These conceptions are, in turn, offered as a scholarly basis for the
very unattractive image of legislation that prevails in modem jurispru-
dence, particularly American jurisprudence.2 7 It is an image that por-
trays legislative activity as deal-making, log-rolling, interest-pandering,
pork-barrelling, horse-trading, and Arrovian cycling-as anything, in
short, except principled political decision-making. We set this sort of
thing up as our model of legislation partly in order to burnish, by
contrast, the reputation of the common law,2 8 and partly, no doubt, to
lend credibility to our normative models of judicial review of legisla-
tion and thus silence misgivings about the undemocratic character of
29
that review.
25. Id. at 13940.

26. For an excellent overview, more balanced than most, see DANIEL A. FARBER &
PHILIP P. FricKEs, LAw AND PUBLIC CHOICE: A CIrrICAL INTRODUCrION (1991).
27. See, e.g., William H. Riker & Barry R. Weingast, ConstitutionalRegulation of Legislative
Choice: The PoliticalConsequences ofJudicialDeference to Legislatures, 74 VA. L. Ray. 373 (1988).
28. This tactic is certainly not a new one. Compare the observations of William Black-
stone, cited in LIEBERmAN, supra note 18, at 56. "For, to say the truth, almost all the per-
plexed questions, almost all the niceties, intricacies, and delays (which have sometimes
disgraced the English, as well as other courts ofjustice) owe their original not to the com-
mon law itself, but to innovations that have been made in it by acts of parliament." Id.
29. Political scientists know better of course. Unlike law professors, they have the good
grace to match a cynical model of legislating with an equally cynical model of appellate
and Supreme Court adjudication. Part of what I am interested in doing in the larger pro-
ject of which this Essay forms a part is to ask: What would it be like to develop, for the
philosophy of law, a rosy picture of legislatures that matched in its normativity and, per-
haps, in its naivete the picture of courts-"the forum of principle"-that we present in the
more elevated moments of our constitutional jurisprudence? See RONALD DWORKIN, A MAT-
TER OF PRINCIPLE 33-71 (1985).
1995] THE DIGNrrY OF LEGISLATION
Our theme in all this is the size of the modem legislature, the
plurality-the sheer numbers-of persons that law-making involves.
How could something which is so obviously a bad idea-in Bagehot's
words law-making by a "big meeting"-have become so entrenched as
a principle of constitutional organization? One response we might
offer is to highlight the efforts that have, in fact, been made in most
legislatures to limit the number of members who participate actively
in the drafting and consideration of a given measure. Legislative pro-
posals may come ready-drafted from the executive, as in Britain, and
their passage controlled by the executive through the mechanisms of
party discipline. Or consideration of a bill may be concentrated
under the much more manageable auspices of a specialist congres-
sional committee, as in the United States. But though these practices
are certainly important, they do little to explain the continuing associ-
ation of the authority of a statute as law with its emergence from an
institution comprising hundreds of representatives who deal with one
another as equals. The requirement that a bill be deliberated upon
and passed by Congress or Parliament as a whole survives, not merely
as a "dignified" charade (like Royal Assent in the United Kingdom),"
but as something which is regarded as a matter of right by the repre-
sentatives themselves and as crucial to the standing and authority of
legislation in the community. Hard-headed political scientists might
locate the real action in the executive, in the corridors of Congress, or
in the committee system. But I have no doubt that if they were advis-
ing constitution-framers in an East European country, for example, on
the appropriate body for making laws, they would urge the institution
of something like a large parliamentary or congressional assembly.
And that would not be understood as a quaint concession to anachro-
nism, like urging them to adopt a constitutional monarchy. It would
be seen as a matter of real importance. Somewhere in our tacit theory
of the authority of legislation is a sense that discussion and validation
by a large assembly of representatives is indispensable to the recogni-
tion of a general measure of principle or policy as law.
II
So far, I have drawn upon a broad variety of materials in what we
might call political, legal, and constitutional theory. One would ex-
pect, however, some more specific and rigorous guidance on these
matters from analytical jurisprudence, more narrowly conceived, par-
30. See BAGEHOT, supra note 24, for the contrast between "dignified" and "effective"
elements in a constitution.
ticularly from positivist jurisprudence, which is reputed to regard de-

liberate law-making-the positing and enforcement of new measures
to govern a community or to change its mode of governance-as the
essence or paradigm of law. One would think that if anyone was to be
interested in the character of modem legislatures, the positivists
would be. In fact, they seem to be quite uninterested. Although posi-
tivist jurisprudence defines law in terms of its sources,"1 its detailed
conceptions of the sources of law-of the institutions that actually
make positive law-remain relatively undeveloped.
Certainly, it is important to most modem positivist theories that
there be legislatures and legislation. H.L.A. Hart thought, for exam-
ple, that the mark of a modem legal system is the community's capac-
ity to deliberately change its rules through formalized procedures.3 2
Hans Kelsen believed that the dynamic role of the Grundnorm con-
tained "nothing but... the authorization of a norm-creating author-
ity" and that in a modem legal order, "the creation . . .of general
legal norms has the character of legislation. "3 But even on this the
positivists are not unanimous. Joseph Raz, who coined the "sources
thesis," 4 has concentrated almost all his institutional attention on
courts, arguing that "the existence of norm-creating institutions
though characteristic of modem legal systems is not a necessary fea-
ture of all legal systems, though the existence of certain types of norm-
applying institutions is."3" Thus on the basis of a strictly theoretical
possibility that a system of positive law might exist without a legisla-
ture, Raz deems any further account of legislation to be inessential for
jurisprudence.
Beyond the bare recognition (by a few) of legislation as a source
of law, what do legal positivists say about law-making institutions?
H.L.A. Hart emphasized that legislatures need not be regarded on the
Austinian model of "uncommanded commanders." A legislature, he
argued, could not be seen as the originating source of all legal rules.
Instead, we must say that a society has a legislature in virtue of the
acceptance in that society of a secondary rule of recognition. We
must understand, moreover, that such a rule may lay down procedural
31. JOSEPH RAz, THE AuTHOrm OF LAw: ESSAYS ON LAw AND MORAirrv 47-48 (1979).
"A law has a source if its contents and existence can be determined without using moral
arguments.... The sources of a law are those facts by virtue of which it is valid and which
identify its content." Id.
32. H.L.A. HART,THE CONCEPT OF LAw 90-93 (1961).
33. HANS KELSEN, THE PuRE THEORY OF LAW 196, 222 (1967).
34. JOSEPH RAZ, THE CONCEPT OF A LEGAL SYsTEM: AN INTRODUCTION TO THE THEORY
OF LEGAL SYSTEM 210-16 (2d ed. 1980).
35. RAZ, supra note 31, at 105; see also id. at 87-88.
1995] THE DIGNITY OF LEGISLATION 643
and substantive norms in the course of qualifying that person or body

to legislate. Now, whether there are such limitations, and if so what
they amount to, is a question to be resolved sociologically on Hart's
account-what rule of recognition is in fact accepted?-not logically
on the basis of any general jurisprudence of sovereignty.3 6 Hart him-
self used the notion that legislatures are constituted by rules to ex-
plain the possibility of composite legislatures (bicamerality, for
example, and the procedures for assent/veto by a head of state) and
the possibility of constitutional limits on legislative power, as in the
United States.
What struck me, however, re-reading The Concept of Law in the
context of this study, was that for Hart these things mattered mainly as
possibilities. The jurisprudential task is to establish that there could
be such constitutive secondary rules. Once that possibility was estab-
lished, Hart seemed largely uninterested in the sort of body that the
modern legislature was. He was personally comfortable using "the
Queen-in-Parliament" as his paradigm. But once the possibilities just
referred were granted, the other points he wanted to make about leg-
islation-about the open texture of language, 7 and about the inter-
play between legislature and courts3 8 -might as easily have been
made with reference to "Rex." Hart has a reputation in America for
being perhaps less interested in courts and in judicial reasoning than
a legal philosopher ought to be. But it is fair to say that he spent
much more time on the issue of who and how many among the polit-
ical elite must accept the rule of recognition that constitutes and em-
powers the legislature, and what the consequences of dissensus in this
group would be, 9 than he spent on the ramifications for jurispru-
dence of the fact that law-making as a unitary act must be constructed
in parliamentary institutions out of a plurality of diverse legislators.
Hans Kelsen was only marginally more forthcoming. In his ac-
count, the idea of constitutive rules of legislative procedure is not just
an abstract possibility: "Only in a domocratic legislation [sic] are reg-
ulations required that determine the legislative procedure, which
here means: participation in the popular assembly or in the election
of the parliament, the number of its members, the proceedings to
pass resolutions. " 4 I take that to mean the following: In a legislature
36. HART, supra note 32, at 69, 144-48.

37. Id. at 121-32.
38. Id. at 138-44.
39. Id. at 59-60, 97-120.
40. KELSEN, supra note 33, at 225. I read "Only in a domocratic legislation" as "Only in
a democratic legislature."
whose composition reflects electoral competition, each elected mem-

ber will claim to be the equal of every other and to have a voice, a vote
and perhaps a mandate that must be heard and registered as a matter
of right. In that context, procedural rules will have an importance
and urgency which they might not have in the consensus politics of an
oligarchy, and which they certainly will not have in a legislative monar-
chy or dictatorship. However, although this point is certainly impor-
tant as a matter of political theory, it was not one that Kelsen
exploited in the rest of his jurisprudence.
In a way, this jurisprudential reticence is more than made up for
by the wealth of empirical descriptions and models of the modern
legislative process that have been developed in political science. As I
have already noted, legal philosophers are not slow to draw on these,
particularly if they present legislatures and legislation in a bad light.
My point then is not that legislatures are suffering from overall aca-
demic neglect, but that, in jurisprudence at any rate, we have not
bothered to develop any idealistic or normative picture of legislation.
Our silence here is deafening compared to our philosophical loquac-
ity on the subject of courts. There is nothing about legislatures or
legislation in modem philosophical jurisprudence remotely compara-
ble to the discussion of decision-making by judges. No one seems to
have seen the need for an ideal type or theoretical model that would
do for our understanding of legislation what, for example, Ronald
Dworkin's "Hercules" purports to do for adjudicative reasoning.4 1
It may be thought that the reason is obvious: Judicial reasoning
poses a special problem for jurisprudence in the way that the reason-
ing of legislators does not. Argument in a legislature is explicitly and
unashamedly political. It is either the interplay of interests, or the
direct clash of policy proposals and ideologies. Legislators do not
need jurists to tell them how to argue. The processes by which courts
reach their decisions, by contrast, are supposed to be special and dis-
tinctive, not directly political, but interpretative of already established
political conclusions or expressive of some underlying spirit of legal-
ity. It is, therefore, a matter of some importance for jurisprudence to
find out whether these claims about the special character of judicial
reasoning can be sustained.
That point can be conceded, but still, jurisprudence needs an
ideal theory of legislation. One has to be careful how one describes
this need. The normative point of an ideal model of legislation would
41. See RONALD DwoRKIN, LAW's EMPIRE 239-40 (1986); RONALD DwORKIN; TAKING
RIGHTS SERIOUSLY 105-30 (1977).
19951 THE DIGNITY OF LEGISLATION
not necessarily be directed at legislators themselves, telling them how

they ought to behave, what laws they ought to pass, what criteria they
should use for the public good, how much attention they should pay
to their constituents' interests as opposed to the interests of the com-
munity as a whole, and so on. The theory we need is one whose les-
sons would be directed at least as much to the legislators' audience-
those to whom their laws are addressed and those by whom they are
interpreted-as to the legislators themselves. In other words, we need
a theory concerning the enactment of positive law that will have
ramifications for how we ought to view such law: how it ought to be
interpreted and what authority it should have in the community.
Think, for example, of the difficulties we have gotten into with
the notion of legislative intention-the idea that in interpreting a stat-
ute we should go back and search the legislative record for guidance.
Whether an appeal to legislative intent in the course of statutory inter-
pretation is either possible or desirable depends in large part on the
sort of institution a legislature is and on the way it acts when it per-
forms its law-making function. Empirical considerations are no doubt
relevant. What evidence of intentions shared by legislators emerges
from the Congressional record or from other sources? To what extent
do legislators actually pursue shared intentions when they legislate?
Realistically, whose intentions get privileged (which legislator? which
sub-group? insiders? freshmen? marginalized legislators? minor par-
ties?) when appeals to legislative intent are made? These empirical
questions are important. But we also need to approach the problem
with a value-laden theory of what legislation by a large diverse body is
supposed to be like, and in what its authority is supposed to consist.
Offhand, it seems that authority is accorded to a law by its enactment;
and since it is the text and not the good intentions of the bill's spon-
sors that are voted on by the legislature, those intentions have no
greater authority than any other sentiments that may be expressed
from time to time in the legislative process. I do not want to argue in
the present Essay for any particular position on this issue,4 2 but I do
want to insist that we cannot address the issue of legislative intention
properly unless we deploy a reasonably detailed model of those as-
pects of the structure, the deliberation, and the decision-making pro-
cedures of legislative institutions that are implicated in a sound theory
of the basis of a statute's authority.
42. But seeJeremy Waldron, Legislative Intent and Unintentional Legislation, in LAw AND
INTERPRETATION (Andrei Marmor, ed.) (forthcoming).
In this regard, our jurisprudence is badly handicapped by a the-

ory of legislation which assumes tacitly that the law maker is a single
person. We talk about legislative intent as though the act of legisla-
tion could be understood using the same model of intentionality that
we use to characterize individual action. And we do this despite our
awareness that modern legislation is enacted by many people with
quite radically varying states of mind and intention.4 3 It is remarka-
ble, for example, that the most recent philosophical defense of ap-
pealing to legislative intent makes its case on the basis of what it
means for a person-one person-to have authority over another, and
then seeks to establish a convention for adapting this view of intent to
the problematic case of an authoritative body comprising hundreds of
members. 4 One would have thought that the wiser strategy would be
to begin by asking how a large number of people acting collectively can
have authority, and then use one's answer to that question to guide
one's approach to what are bound to be, on any account, difficult and
controversial issues about legislative intent.45
I have argued that philosophers of law, in the positivist tradition,
are largely indifferent to the composition of legislatures: They give
the impression that we might as well treat the legislature on the model
of a single individual, for all the difference that it makes to jurispru-
dence. In fairness, I should add that in the seminal works of modern
positivism-Jeremy Bentham's work, for example, and John Aus-
tin's-one does find a gesture towards the idea (which of course was
the political reality in contemporary England) that a legislature might
be a large and numerous body. Bentham said neutrally that we iden-
tify a sovereign whenever we notice "any person or assemblage of persons
to whose will a whole political community are (no matter on what
account) supposed to be in a disposition to pay obedience: and that
in preference to the will of any other person."4 6 However, that
phrase-"assemblage of persons"-was almost his only concession to
the point. For the rest of his jurisprudence and in much of his polit-
43. For a the critique of the intentionalist approach, see, for example, DWORKIN, LAW'S
EMPIRE, supra note 41, at 317-21.
44. ANDREI MARMOR, INTERPRETATION AND LEGAL THEORY 113-23, 159-65 (1992); see also
JOSEPH RAZ, THE MORALYrr OF FREEDOM 23-105 (1986).
45. See Waldron, supra note 42.
46. JEREMY BENTHAM, OF LAws IN GENERAL 18 (H.L.A. Hart ed., Athlone Press 1970)
(1782) (emphasis added). Austin's language is similar: "Every positive law, or every law
simply and strictly so called, is set by a sovereign person, or a sovereign body of persons, to
a member or members of the independent political society wherein that person or body is
sovereign or supreme." I JOHN AUSTIN, LECTURES ONJURISPRUDENCE 116 (Robert Campbell
ed., 1874) (Lecture VI).
1995] THE DIGNrIY OF LEGISLATION 647
ical philosophy, the sovereign was almost always referred to as "he."47

In the Introduction to the Principles of Morals and Legislation, the implicit
recipient of Bentham's advice almost always appears in the text as "the
legislator," with no reference at all to the pluralistic possibility.
One way to read this is to see it as indicative sub rosa of a definite
bias in political philosophy in favor of monarchy-whether in the
form of the philosopher-king, the unitary sovereign, or the enlight-
ened despot. Certainly, that is what one would conclude from associ-
ating the positivism of Bentham and Austin with that of Thomas
Hobbes. For when Hobbes said "the Legislator is he ... ,"48 the pro-
noun was not at all neutral as to number, but reflective of his personal
conviction that monarchy was by far the best form of government,
among other reasons because "a Monarch cannot disagree with him-
selfe, out of envy, or interest; but an Assembly may; and that to such a
height, as may produce a Civill Warre."49 All the same, because Hob-
bes had no choice but to acknowledge that this "one thing alone I
confesse in this whole book not to be demonstrated, but only probably
stated,"5" the strict logic of his position required him to repeat from
time to time that, in theory at least, sovereignty might be vested
"either in one Man, or in an Assembly of more than one.
It would be wrong, however, to think of modern legal positivists
as closet monarchists. Their persistence with the image of legislation
as analogous to one man's act of will stems rather, I suspect, from an
implicit belief that the structure and character of a legislature is
largely uninteresting from the point of view of legal philosophy.
Although, as I stated earlier, positivism defines law in terms of its
sources, the property of being a source of law is, for the positivist, about
the most interesting thing that can be said about a legislature, a legis-
lator, or a legislative institution. Who cares whether a Parliament has
one house or two, 435 seats or a hundred? We can treat it as a black
box. What matters is that this is where laws come from, this is the
agency by which they are posited, and that they have the status of law
in virtue of this origin rather than in virtue of the moral character of
their contents. Representing the legislature, at least stylistically, as
47. I am not saying that Bentham was uninterested in legislative assemblies. On the
contrary, he can claim to have founded the peculiar English obsession with the shape and
furniture of legislative chambers-an obsession which, for example, surfaces again in the
writings and speeches of Winston Churchill. See K.C. WHEARE, LEGISLATUREs 7-13 (1963).
But Bentham saw no consequences for jurisprudence in the assembly aspect of legislation.
48. THOMAS HOBBES, LEVIATHAN 185 (Richard Tuck ed., Cambridge 1991) (1651).
49. Id. at 132.
50. THOMAS HOBBES, DE CrvE 37 (Howard Warrender ed., Oxford 1983) (1651).
51. See, e.g., HOBBES, supra note 48, at 120, 129, and 184.
though it had the character of a single author of the texts we call

statutes is a logical representation of the singularity of the source qua
source. It is not, I think, intended as a politically-loaded or wishful
representation of the ideal character of the source-institution itself.
The point is connected, in a way, with what positivists do in mod-
em analytical jurisprudence. They spend most of their time defend-
ing the general idea of the sources thesis against proponents of
various forms of natural law. In that debate, what matters is the possi-
bility that a community might confer the status and authority of law
on certain norms by virtue of their institutional provenance. What
matters is not one account of the institutional details versus another,
but rather the abstract idea of institutional provenance as the basis of
law's authority rather than substantive criteria of morality or justice.
Once that abstract idea is established, the institutional focus shifts very
quickly to the courts. This is because modem natural lawyers are
much less interested these days in defending the proposition lex ini-
usta non est lex against the sources thesis,5 2 than they are in arguing
that the sources thesis cannot give an adequate account of what goes
on in the courtroom.
III
There is one additional explanation for our neglect of the com-
position and processes of legislatures in modem legal philosophy. It
has to do with the level of generality at which analytical jurisprudence
defines its agenda. Its practitioners believe that there are important
philosophical tasks to be performed at the level of general jurispru-
dence-that is, jurisprudence addressing the very ideas of law and
legal system, apart from the peculiarities of particular jurisdictions.5 3
From the perspective of general jurisprudence, the possibility that a
legislature might consist of one person or many persons is a variable
that is simply beneath notice. Because those who take this approach
do not want general jurisprudence to confine itself to the study of
democracies, for example, they hold that we should not build into our
concept of legislation the premise that it is necessarily the product of
52. Cf JOHN FiNis, NATURAL LAw AND NATURAL RiGH-TS 23-49 (1980).
53. We see this approach taken to an extreme in Joseph Raz'sjurisprudence, where on
the basis of the mere logical possibility that a legal system might exist without a legislature,
Raz deems an investigation of legislatures to be philosophically inessential. See RAz, supra
note 34 and accompanying text. Raz takes a similar view about the relation between law
and sanctions: since a legal system could exist without sanctions, the fact that all existing
legal systems uphold their norms with rules and penalties is a mere contingency, and thus
of limited philosophical interest. Id, at 175-86.
the many acting in a representative assembly. We want our philoso-

phy of law, they will say, to tell us what is in common between legisla-
tion by a king in Saudi Arabia and legislation in the American
Congress or the German Bundestag.
I suspect this quest for institutional neutrality in legal theory is
largely misguided. Of course, if we stipulate a sufficiently high level of
abstraction for our general jurisprudence, we will have no choice but
to ignore the features that distinguish some sources of law from
others. We might, for example, try to find a definition of positive law
that blurred the distinction between statute and custom, or a defini-
tion of "source of law" that licensed no distinction between legislation
by a parliament and judge-made law in a hard case. But after a while,
the payoffs would begin to evaporate in the heady realms of such ab-
straction, and we would be overwhelmed by the distortions introduced
by a theory that insisted one size fits all.
In fact, of course, the modern positivist notion of source-of-law is
not institutionally neutral. One way of seeing this is to consider the
difference that was made in jurisprudence when the paradigm of man-
made law shifted from custom to legislation. Customs can be seen as
immemorial, earthy, local, and conventional, inhabiting a region be-
tween habit, prescriptivity and consent that is only partially captured
by modem notions such as the "internal aspect" of rules.54 They need
not be mindless artifacts of common behavior. On the contrary, ju-
rists have always talked of customs being held, remembered, and in
that sense posited in a community: "According to the ancients, a local
custom is an establishment held and preserved in a country by the old
wise men by agreement, and maintained according to the condition
of the place as long as it is accepted and suffices."5 5 What is more,
customs may even be worked up into an articulated system, capable of
intellectual coherence and relatively conscious growth; we all know
that as part of the historical ideology of the common law, whether or
not it corresponds to reality. For generations ofjurists, particularly in
Western Europe, custom was the soul of man-made law.
Now, there is a world of difference-marked by a century or two
of savage disagreement among lawyers-between this jurisprudence
of custom and a legal philosophy that identifies man-made law as typi-
cally the expression of a legislator's sovereign will. In a way .that is
perhaps difficult for us to grasp, the partisans of custom in the eight-
54. See HART, supra note 32, at 54-56, 83-88.

55. JEAN BOUTELLIER, SOMME RuRALE (1603), cited in DONALD KELLEY, THE HUMAN MEA-
suRE: SocIAL THOUGHT IN THE WESTERN LEGAL TRADION 104 (1990).
eenth and nineteenth centuries regarded legislation-by an enlight-

ened despot, for example-as having much more in common with the
scientistic "Natural Law" of the philosophers and the theologians than
with the humble sublunar reality of the norms which generations of
ordinary folk had posited for themselves in various countries. Legisla-
tion tended to be associated with reason and rational thinking, with
the ability of one enlightened mind to cut through the accreted layers
of local usage.5 6 Where custom was inarticulate, legislation presented
itself in shining formulas of principle. Where custom was irredeema-
bly parochial, legislation might be the same everywhere-reason's
common solution to common human problems. And where custom
was consensual, legislation could be the coercive off-spring of sover-
eignty-not inherently oppressive, of course, but something that
might have to be imposed on an ignorant people by one more enlight-
ened than they.
There is no space to pursue this contrast any further in the pres-
ent Essay. 57 But no one can have any doubt about which conception
has triumphed in modem positivist jurisprudence-particularly legal
positivism as it has developed out of the English school of Bentham
and Austin. There is a massive difference-in substance, ethos and
rhetoric-between a positivist jurisprudence dominated by an image
of customary law and a positivist jurisprudence dominated by an im-
age of legislator's law. The hegemony of the latter has perhaps re-
treated a little in Anglophone jurisprudence in the last thirty years.
We are well aware of the attempts that were made, in behalf of the
legislator model to subsume custom altogether-for example, in Ben-
tham's definition of customary law as "a miscellaneous branch of stat-
ute law ill-expressed and ill-defined."5 ' And we know, from Hart's
work among others,5 9 how to resist that assimilation. We are now not
fooled by the alleged neutrality of the legislator model into inferring,
from the fact that both custom and legislation count as positive law,
that they are therefore both the product of someone's will.60
56. Cf PHILIP SELZNICK, THE MORAL COMMONWEALTH: SociAL THEORY AND THE PROM-
ISE OF COMMUNrY 270 (1992) ("Rational authority has quite a different foundation [than
traditional authority]. Instead of immemorial custom, whose origins are lost in mystery,
some form of enactment is the font of legitimacy. Rationality calls for explicitness, coher-
ence, predictability, and impersonality.").
57. See KELLEY, supra note 55, at 91-108 and 167-206, for a rich, detailed and nuanced
account of the role of custom in European jurisprudence, from the twelfth to the seven-
teenth century.
58. BENTHAM, supra note 46, at 235.
59. HART, supra note 32, at 44-48, 77-96.
60. There is an excellent brief discussion of some differences in KELSEN, supra note 33,
at 224-29.
In much the same way, I think we should be open to the possibil-

ity that there is a similar discrepancy, similarly relevant for jurispru-
dence but belied by the common term "legislation," between the
positing of laws by one enlightened legislator and the law-generating
activity of a large-scale representative assembly involving hundreds of
members. We should certainly be alert to the consequences of contin-
uing to use the rhetoric and ethos of a single-legislator model to de-
scribe a practice which is structurally quite different. It is different,
for example, because it involves more than one person, and thus nec-
essarily involves relations among persons in a way that single legislator
models do not.6 ' We should be particularly alert to this danger, since
we know that the single-legislator model we are using-as our "neu-
tral" model of legislation-is one that has already tried and failed to
render reductively in its own image other diverse, and importantly di-
verse, sources of positive law.
IV
In the remainder of this Essay, I want to explore one way in which
taking the composition of legislatures seriously might have important
pay-offs for our understanding of statutes and of the way they present
themselves to us in a modem legal system. I shall pursue a hunch
about a connection between the diverse composition of legislatures
and something which is sometimes referred to as the peculiar "textual-
ity" of statutes.
Legal theorists have long been fascinated by the textual quality of
law-by the fact that legal conundrums can often be expressed as
problems about the meanings of words.62 I doubt whether "the letter
of the law" is an idea associated only with legislation; custom has its
"letter" too, as it signifies its presence in well-worn, almost incantatory
sayings. But legislation is the legal form in which emphasis on the
words themselves-ipsissima verba-is most prominent in modern ju-
risprudence. A statute consists of an enacted form of words; and the
61. Compare the criticisms of utilitarianism as a basis for political philosophy in JOHN
RAwLs, A THEORY OFJuSTICE 28-29 (1971):
There is no reason to suppose that the principles which should regulate an associ-
ation of men is simply an extension of the principle of choice for one man. On
the contrary- if we assume that the correct regulative principle for anything de-
pends on the nature of that thing, and that the plurality of distinct persons with
separate systems of ends is an essential feature of human societies, we should not
expect the principles of social choice to be utilitarian.
62. See, e.g., LEGAL HERMENEUTICS: HISTORY, THEORY, AND PRACTrICE (Gregory Leyh ed.,
1992); CosTAs DouziN.As ET AL., POSTMODERN JURISPRUDENCE: THE LAW OF TExT IN THE
TEXTS OF LAw (1991).
652 MARYLAND LAW REviEW [VOL. 54:633
problem of interpreting and applying a statute is usually presented as

the semantic problem of bringing a given form of words into relation
with persons, things and events in the world.
Atiyah and Robert Summers have remarked that one could imag-
ine a legal system according "no particular respect to the verbal form
in which legislative texts are cast, treating statutes simply as cases are
treated in the common law system." 6" In fact, that is not how we treat
statutes. We treat pieces of legislation as texts: We feel ourselves
bound by and associate authority with the very words that the legisla-
ture produces to a much greater extent than we do in regard to other
sources of law or, for that matter, other sources of authority. Unlike
common law principles, for example, the standards embodied in stat-
utes have canonical formulations: The rule that has been posited by
the legislature is the rule expressed by the very words that are used in
the bill that the legislature has passed.
This contrast between legislation and other sources of law may be
misunderstood. There are aspects of textuality to almost all modern
law. In a common law system, for example, there are authoritative
renderings of the texts of judgments handed down by courts, includ-
ing the statements of reasons offered by each judge. There are, in
other words, official law reports, which effectively settle any dispute
about what was said in giving judgment. Even so, in the common law
tradition, what the judge said (in the sense of the exact words that
were used) is not taken to be essential to the identity of whatever stan-
dard was laid down. A different judge may be thought to be following
the same principle even though she uses different words; and when
she does, there is no sense that the original wording is canonical and
hers a revision or reinterpretation. The principle simply does not
have a canonical expression.' Interpretive exercises in common law
are thus not oriented towards a given form of words as they are in the
65
case of statutes.
63. PATRICK S. ATIYAH & ROBERT S. SUMMERS, FoRM AND SUBSTANCE IN ANGLO-AMERICAN
LAW 97 (1987).
64. But see KENT GREENWALT, LAW AND OBJEcTrVrrv 66 (1992), for a discussion of the
idea of "overlapping" formulations.
65. This is not to be read as implying that the distinction between canonically-worded
and non-canonically-worded standards corresponds to the distinction between "rules" and
"principles" as in Ronald Dworkin's earlyjurisprudence. See DwogRIN, TAKING R cHs SERI.
OUSLY, supra note 41, at 22-31. Dworkin's distinction turns, not on textual canonicity, but
on the kind of normative force that different standards are conceived to have. Id.
Dworkin insists, moreover, that it is a mistake to suppose that even statutory rules have
canonical formulations: "[1]t is a commonplace that lawyers will often misrepresent the
rules that a statute enacted if they simply repeat the language that the statute used. Two
lawyers might summarize the effect of a particular statute using different words... ; they
To say this is not to deny that statutes need interpretation or that

the words of the enactment (and their "plain meaning") are often
insufficient to determine the statute's application. The point is rather
that, in the case of statutes, the problem of interpretation begins from
a sense that there is "a single, definitive, linguistic formulation"" of
the standard being considered-there are words there which may (or
may not) actually have a plain meaning. In the case of other sources
of law, hermeneutical difficulties get going on a somewhat different
basis.
There are of course important differences between legal systems
in this regard. Atiyah and Summers note that English judges "tend to
adopt a more textual, literal approach, while American courts tend to
take a more purposive and, therefore, substantive, approach" to stat-
utes.6 7 But it is possible to exaggerate these differences,' and in any
case, the point is not that statutory language is treated with exactly the
same respect the world over, but rather that it is treated, in every legal
system, with somewhat more focused respect than the language used
by other sources of law in the same legal system (including other tex-
tual sources of law).69
How, then, are we to explain the peculiar textuality of statutes? It
is not enough simply to say that the legislature is an authoritative
source of law, and that this is why its ipsissima verba must be respected.
For we do not treat all authorities in this way-hanging, as it were, on
their every word. My hunch is that the answer has to do with the sort
might still both be saying the same thing." Id. at 76. Of course, even if this were true, i.e.,
that standards are constructed through interpretation, it may still be the case that the
language used in the statute is canonical in regard to the process of interpretation in a way
that the language used by other sources of law is not.
I doubt that Dworkin would deny this. His recent characterization of the judge's task
in regard to legislation is as follows: "He [i.e., Hercules, the ideal judge] tries to show a
piece of social history-the story of a democratically elected legislature enacting a particu-
/ar text in particular circumstances-in the best light overall...." DwoRKN, LAw's EMPIRE,
supra note 41, at 338 (emphasis added).
66. See Robert S. Summers, Statutes and Contracts as Founts of FormalReasoning, in EsAs
FOR PATRICK ATrYAH 71, 74 (Peter Cane and Jane Stapleton eds., 1991).
67. AmA M & SUMMERS, supra note 63, at 100-01.
68. Atiyah and Summers cite an observation byJudge Posner, Statutoy Intepretation-in
the Classroom and in the Courtroom, 50 U. CHS. L. Rxv. 800, 807-08 (1983), to the effect that a
judge "rarely starts his inquiry with the words of the statute, and often if the truth be told,
he does not look at the words at all." Posner himself acknowledges that this is more accu-
rate as a description of constitutional interpretation, and that it applies to legislation only
in regard to statutes which have been around for a long time, which have accumulated a
large body of case law around them, and which even in their original wording invoked
common law concepts and traditions; for example, the Sherman Act. Id. at 808.
69. See, for example, the comparison between statutes and contracts in American law,
in AmviYAH & SUMMERS, supra note 63, at 42.
of institution a legislature is: a large gathering of disparate individuals

who purport to act collectively in the name of the whole community,
but who can never be sure exactly what it is that they have settled on,
as a collective body, except by reference to a given form of words in
front of them. This is the argument that I want to develop.
V
I began with the familiar fact that everywhere legislatures consist
of hundreds of members. These throngs of people do not simply as-
semble from time to time and vote on laws. It is surely part of our
normative image of legislatures that they are bodies where delibera-
tion takes place. Think back to the John Adams's idea outlined in
Part I: A representative legislature "should be an exact portrait, in
miniature, of the people at large,"7 ° so that when the legislators talk to
one another, different parts of society can be taken, through their
representation, to be talking to one another.
The importance of deliberation was also emphasized by John Stu-
art Mill:
Representative assemblies are often taunted by their enemies
with being places of mere talk and bavardage. There has sel-
dom been more misplaced derision. I know not how a repre-
sentative assembly can more usefully employ itself than in
talk, when the subject of talk is the great public interests of
the country, and every sentence of it represents the opinion
either of some important body of persons in the nation, or of
an individual in whom some such body have reposed their
confidence. A place where every interest and shade of opin-
ion in the country can have its cause even passionately
pleaded, in the face of the government and of all other inter-
ests and opinions can compel them to listen and either com-
ply or state clearly why they do not, is in itself, if it answered
no other purpose, one of the most 71
important political insti-
tutions that can exist anywhere.
But we saw earlier that Mill did not entertain a particularly rosy view of
parliaments as places for action.7 2 It is not clear that he regarded this
business of "talking" as anything other than an expressive opportunity
to vent a variety of grievances and complaints.
There is, however, an older tradition in political theory which as-
signs an important synthetic function to talking in the context of prac-
70. PrTKIN, supra note 10, and accompanying text.

71. Mira., supra note 21, at 82-83.
72. See supra note 21, and accompanying text.
tical deliberation. That tradition has its most explicit origin in the
Politics of Aristotle. For, although Aristotle inclined to the view that
the law-giver in an ideal society would be "the one best man,"7 3 he
conceded that
There is this to be said for the Many. Each of them by him-
self may not be of a good quality; but when they all come
together it is possible that they may surpass-collectively and
as a body, although not individually-the quality of the few
best. Feasts to which many contribute may excel those pro-
vided at one man's expense. In the same way, when there
are many [who contribute to the process of deliberation],
each can bring his share of goodness and moral prudence;
and when all meet together the people may thus become
something in the nature of a single person, who-as he has
many feet, many hands, and many senses-may also have
many qualities of character and intelligence. This is the rea-
son why the Many are also better judges [than the few] of
music and the writings of poets: some appreciate one part,
some another, and all together appreciate all.74
What lies behind this is the idea that a number of individuals may
bring a diversity of perspectives to bear on issues under consideration,
and that they are capable of pooling these perspectives to come up
with better decisions than any one of them could make on his own.
That, after all, is why Aristotle took it as the mark of man's political
nature that he was endowed with the faculty of speech.7 ' Each can
communicate to another experiences and insights that complement,
complicate or qualify those that the other already possesses; and when
this happens in the deliberations of an assembly, it enables the group
as a whole to attain a degree of practical knowledge that surpasses
even the coherently applied expertise of the one excellent legislator.
We may or may not buy Aristotle's view that the many can in this
way come up with better results than the one. But the existence of
diverse perspectives in the community and the helpfulness of bringing
them to bear on proposed laws are surely important features in any
account of why the task of legislating is entrusted to assemblies. I be-
lieve that 'these features in turn frame the way in which we should
think about the deliberative process itself, and in particular how we
73. THE PoLrrxcs oF AiusoTLE 141 (Ernest Barker trans., Oxford 1958) (bk. III, ch. xv,
1286a).
74. Id. at 123 (bk. III, ch. xi, 1281a-1281b) (passages in square brackets are the transla-
tor's interpolations.) See also id. at 141-42 (bk. III, ch. xv, 1286a).
75. Id. at 5-6 (bk. I, ch. ii, 1253a).
should regard the relatively high level of 76 formality associated with de-
bate and action in a legislative assembly.
I said a moment ago that Aristotle's view about the wisdom of the
multitude is connected intimately with his teaching that the mark of
man's political nature is his capacity for reasoned speech. It is worth
pausing at this point to notice a renewed emphasis on speech, dis-
course and conversation in political and legal theory, particularly in
theory that has been influenced by the work of Jurgen Habermas.77
Jurists influenced by Habermas have suggested, for example, that a
commitment to the United States Constitution is "less a series of pro-
positional utterances than a commitment to taking political conversa-
tion seriously.... [T]he Constitution is best understood as supportive
of such conversations and requiring a government committed to their
maintenance.",7 They suggest too that the authority of judicial deci-
sion-making is "sustained less by the perceived correctness of the re-
sult than by the coherence of the result, for coherence permits
conversation, and it is ultimately
79
a faith in our ability to converse mor-
ally that holds us together."
Much of this is far-fetched in ways I cannot go into here-in its
aestheticism, for example, or in its conception of discourse as an end
in itself. But one point is particularly important for our purposes.
There is a constant temptation in modern discourse-jurisprudence to
take as an implicit procedural ideal the model of an informal intimate
conversation among friends. Certainly, an informal conversation
among friends has attractive features of equality, openness and mu-
tual respect. But it also tends to be predicated upon the idea that
participants share implicit understandings and that their interaction is
oriented towards the avoidance of adversarial disagreement and the
achievement of consensus. These, I think, are qualities which are
76. In their discussion of the inherent formality of statute law, Atiyah and Summers
neglect this aspect. See supra note 63, 96-114. They stress (i) its "rank formality" (i.e.,
priority over other forms of law), (ii) its "content formality" (i.e., tendency to embody
arbitrary or conventional elements), (iii) its "mandatory formality" (i.e., the application of
standards embodied in a statute is non-optional; it cannot be "distinguished" as case-law
can), and (iv) its "interpretive or textual formality" (i.e., the identification of the law em-
bodied in a statute with a canonical form of words). The aspect on which I am concentrat-
ing here may be regarded as a fifth dimension: (v) procedural formality. I shall argue in
Part VI, infra, that (v) contributes considerably to our understanding of (iv).
77. See especiaIlyJURGEN HABERMAS, MORAL CONSCIOUSNESS AND COMMUNICATIVE ACTION
(1990).
78. SANFORD LEVINSON, CONSTITUTIONAL FAITH 193 (1988).
79. LrEF H. CARTER, CONTEMPORARY CONSTITUTIONAL LAW-MAKING: THE SUPREME
COURT AND THE ART OF POLITICS 143 (1985).
quite misleading so far as our models of political deliberation are

concerned.80
Suppose, then, that instead of a conversation among friends we
were to take the following as our ideal type of legislative deliberation.
A large number of persons have assembled in a hall as representatives
from different parts of a diverse society. Let us suppose that it is a
radically diverse society-so that the members of the assembly repre-
sent not only different interests and regions, but come from com-
pletely different backgrounds, ethnic and cultural, as well as whatever
political differences divide them. Imagine, for example, the national
legislature in India or some other vastly diverse modern state. The
representatives may belong to various religious traditions; they may be
familiar with quite different social forms; they may have disparate
senses of what gives meaning to life. They may not even speak the
same language. Perhaps there is a state language stipulated for their
proceedings in the legislature; if so, we may think of it as a second
language for most of them, and one that they must use carefully and
hesitantly. Certainly, their presence together in the chamber attests
to the fact that they share some sense of common purpose-though it
may not be much more than a foreboding that any attempt to disen-
tangle their diverse interests into more homogenous "nation"-states,
each with a legislature of its own whose members really do understand
each other, would be fraught with the most frightful dangers and
difficulties.8 1
To the extent that such a body seeks to legislate on common
problems, there is a need for the members to be able to talk to one
another, so that each can contribute insights and perspectives that
would otherwise be quite outside the experience of the other legisla-
tors. However, the very reasons that make this interaction desirable
and necessary also make it quite unlikely that the members can pro-
ceed with their deliberations as though they were conducting an
open-ended conversation among friends. They are not transparent to
one another as friends are, and they do not have a great deal of com-
mon ground on which confidences could be shared, premises as-
sumed, and nuances taken for granted. Indeed, the prospects for
mutual misunderstanding and for talking at cross-purposes are greatly
80. Even among friends, conversation may have to become more formal to the extent
that it is purposive and action-directed rather than casual. I am grateful to Philip Selznick
for this point.
81. See the discussion of Salman Rushdie's apprehensions about communitarian poli-
tics in India, in Jeremy Waldron, Minority Cultures and the Cosmopolitan Alternative, 25 U.
MICH. J.L. REFo RM 751, 792-93 (1992).
MARYLAND LAW REVIEW [VOL. 54:633
enhanced by the very features of their situation that make it impor-

tant-on Aristotelian grounds-for them to talk to one another.
These representatives cannot deal with one another in the way that
members of a tightly-knit gemeinschaft or an "old-boy's network" are
often thought to be able to deal with one another. They share very
little beyond an overlapping sense of common problems, and the
rather stiff and formal language that they address to one another in
their debates about those problems. If any one of them says, in the
rather cozy way that people have who share tacit understandings,
"Come on, you know what I mean," the answer is as likely as not to be:
"No, I don't know what you mean. You had better spell it out for me."
Of course this is an extreme example. The membership of the
United States Congress is less diverse than this, and the British House
of Commons is more homogenous still. There are reasons for dwell-
ing on the more extreme model nonetheless.
First, as we have seen, the less diverse the body of representatives,
the weaker the case for legislation by an assembly as opposed to legis-
lation by a single representative individual. Both John Adams's case,
the "microcosm" model, and the Aristotelian case for legislation by
the multitude rely on the fact that different members will bring to the
process differing perspectives and experiences.
Second, and more generally, exaggeration offers certain method-
ological advantages in ideal type analysis.8 2 By highlighting the diver-
sity associated with the sheer plurality of a modern legislature, the
model I am framing may help us understand the challenges and tech-
niques that come along with diversity, so far as deliberation and deci-
sion-making are concerned. True, in a more homogenous legislature,
these concerns can be discounted pro tanto. There is, nevertheless,
some theoretical advantage in using an exaggerated model to focus
our attention on diversity among legislators, particularly when-as I
shall argue-they offer a plausible basis for understanding certain fea-
tures of modem legislation which are otherwise somewhat bewilder-
ing: for example, the high degree of formality associated both with
legislative deliberations and with legislative outputs.
Third, there are normative reasons for taking this approach. We
pride ourselves on having not only a diverse society, but a political
philosophy-particularly liberal political philosophy-oriented specif-
82. See 1 MAX WEBER, ECONOMY AND SOCIETY 19-21 (Guenther Roth & Claus Wittich
eds., Fischaff et al. trans., U. of Cal. Press 1978).
ically to the challenges and difficulties of that diversity.8 3 Analytic ju-

risprudence should participate in that orientation, while in its
philosophical assumptions about the sources of law, it should not err
on the side of social homogeneity.8 4 Even if the diversity which we
actually face is somewhat less extreme, the liberal instinct is that we
should not be counting on that. Indeed, it is an important criticism of
recent communitarian theories that they implicitly presuppose and
rely upon on a degree of cultural and ethical homogeneity to an ex-
tent that is really quite reckless in the circumstances of the modern
world.
We should be careful, in other words, to avoid building in any
premise of ethnic and cultural homogeneity as a prerequisite in our
models of politics and legislation. The narrow and intolerant im-
pulses of communitarianism, tribalism and nationalism are wreaking
enough havoc in the world as it is, without the encouragement of lib-
eral jurisprudence. If there is a conception of law that makes no such
assumption-or that celebrates diversity or, at any rate, comes to
terms with it-we should try to state that conception clearly and make
it central in our jurisprudence if we can.
VI
We left our diverse legislators in their chamber, a little unsure of
each other and a little tongue-tied. We might imagine that from time
to time one of them will stand up and make a proposal to her fellow
representatives. When she does, it may evoke a flurry of responses.
Some she will recognize as opposing her idea; some will appear to
misunderstand it. Others may put forward a different proposal or a
counter-proposal, which in turn will evoke similarly confusing re-
sponses. Some members may stand up and make speeches on quite
different matters. They may tell stories of tenuous or indeterminate
relevance about the problems and experiences of their constituents.
Others may interrupt by chanting slogans, singing hymns, telling
jokes, or shouting threats. Matters that various members conceive as
having great urgency will cut across one another. Responses to one
idea will be taken as responses to a different idea, and no one will be
able to keep track of where they have got to on any particular front.
83. See especiallyJoHN RAWLS, PoLrrIcAL LIBERALISM (1993). For a powerful argument
that Rawlsian liberalism does not go far enough in addressing diversity and difference, see
IRs MARION YOUNG, JUSTICE AND THE POLmCS or DIFFERENCE (1990).
84. See MARTHA MINow, MAKING ALL THE DIFFERENCE: INCLUSION, EXCLUSION AND
AMERICAN LAW (1990).
MARYLAND LAW RE [VOL..w54:633
In circumstances of this kind, we see the importance of delibera-

tive formality, parliamentary procedures, and rules of order. It is clear
from the scenario just outlined that an assembly like the one we are
imagining needs to structure and order its deliberations if it is to
achieve any of the advantages that the Aristotelian theory suggests may
accrue from legislation by the many. The members of our assembly
need to establish rules and procedures that address issues such as the
following: How are debates initiated and how are they concluded?
Who has the right to speak when, how often, and for how long? Who
may interrupt, who may exact an answer to a question, who has a right
of reply? How is a common sense of relevance maintained or how are
members assured that they are not talking at cross purposes? What
issues, subjects, or details may be addressed at various stages in the
proceedings? How are topics for debate selected, how are subject-mat-
ter priorities set, and how is an agenda determined? How is the con-
duct and conclusion of a deliberative session related to the assembly's
powers of resolution and action? How is debate brought to an end?
How are decisions taken?
Blandly stated, these matters might seem obvious and beneath
notice in jurisprudence. It is tempting to regard such procedural
rules as arbitrary conventions, with no intrinsic philosophical signifi-
cance of their own. That would be a mistake. Political scientists have
noted the remarkable similarity in parliamentary procedures around
the world.8 5 There are no doubt historical reasons: for example, the
global influence of the Westminster tradition, largely as a legacy of
British imperialism. But the similarity can also be understood as a
common human response to a similar set of problems: the circum-
stances of procedure, so to speak.8 6 Wherever there is a felt need for
common action as an upshot of deliberation in circumstances of diver-
sity and disagreement, then needs of the kind outlined in the previous
paragraph are bound to be felt.
VII
I have focused so far on what the ideal model of a diverse legisla-
ture contributes to our understanding of procedural formality in law-
making. I now want to turn to the relative formality of legislation-
i.e., statutes-considered as products of processes of this kind.
85. See, e.g., PHILIP LAUNDY, PARUAMENTS IN THE MODERN WORLD 61 (1989).
86. Cf RAwLs, supra note 61, at 126-30 (describing co-existence, mutual vulnerability,
moderate scarcity, limited altruism, and ethical and religious diversity as "the circum-
stances of justice").
My hunch is that textual canonicity and the procedural formality

discussed in the previous parts are connected. At first glance, how-
ever, there is no obvious reason that procedural formality should nec-
essarily issue in output-or interpretive-formality. We might
imagine a proclamation made by a legislator-monarch with all sorts of
formal pomp and ceremony, and yet it might be understood that the
purpose of the occasion's formality is to impress upon subjects the
personal authority of the ruler rather than to evince any commitment
on his part to the particular linguistic expressions that were used. In-
deed, it might follow from the fact that the procedural pomp high-
lights the personal authority of the monarch, that we should orient
our behavior to what we think he intended rather than the words he
happened to say. In these circumstances, one might want to say, with
Hobbes, that "it is not the Letter, but the Intendment, or Meaning...
(which is the sense of the Legislator,) in which the nature of the Law
consisteth." T By itself, then, procedural formality is not sufficient to
make sense of output formality.
It is possible, however, to relate output formality to procedural
formality, when the later is undertaken for certain reasons. Think
back to the predicament of the radically diverse legislature we
imagined in Part VI. The members of that body found it necessary to
order their deliberations with formal rules governing the setting of an
agenda, the initiation and conclusion of discussion, the right to speak,
the structure of debate, and the basis of voting and decision-making.
Those familiar with parliamentary procedure, whether in national leg-
islatures or small-scale public meetings, will know that debating rules
are largely oriented towards and ordered by the idea that at any given
time there is a specified proposition under discussion. Even in the
most basic forums, propositions are moved and seconded, and once a
motion is on the table, the deliberative body has a specific form of
words which is open for discussion, and the organizing principle of
debate becomes (roughly) that all and only contributions to the con-
sideration of that proposition are to be heard, until either the time set
for debate runs out or until the question defined by that proposition
has been resolved. This insistence that there be a formulated motion,
and that speakers observe norms of relevance in regard to that mo-
tion, are the primary basis on which parliamentary procedure seeks to
avoid the nightmare of people talking endlessly at cross-purposes and
the failure to make the sort of contact with one another's contribu-
tions (synthetic or dialectic) that practical deliberation requires.
87. HOBBES, supra note 48, at 190.

MARYLAND LAW REI EW [VOL. 54:633
No doubt, in the course of discussion, someone may feel that it

would be wiser for the assembly to discuss a somewhat different propo-
sition than the one specified, perhaps worded in a subtly or substan-
tially different way. But if they want to press the point, the
parliamentary rule is that they must move an amendment, changing
the wording of the motion under discussion, once again in a specifi-
cally formulated way. Proceedings are then devoted to a discussion of
the virtues of the amendment, qua amendment, and a vote is taken on
that, before the substantive discussion is resumed. And again, we see
the virtue of this way of doing things in a diverse assembly. In conver-
sation among friends, the topic may shift in an open-ended way, and
people familiar with one another have both the willingness and the
ability to keep track. But in an assembly consisting of people who are
largely strangers to one another, deliberation would be hopeless if
there was a sense that the topic might or might not have shifted
slightly after every contribution. So, although amendment processes
exist, their formulaic character and the rules governing their proposal
and adoption provide a way of keeping track of where the discussion
is, a way of keeping track which does not depend upon implicit under-
standings that some of the members may not share.
When discussion is exhausted, a vote may be called for, and-if
my experience of law faculty meetings is any indication-someone will
immediately leap to their feet and say: "I'm confused. What exactly
are we voting on?" In a well-run assembly, the clerk or secretary will
be in a position at that stage to read out the proposition (as amended)
which now is the focus of the final vote. Once again, the determinacy
of that proposition, as formulated and as amended, is important to
establish a sense that we are all orienting our actions in voting to the
same object. It is important for me to know, for example, that what I
take myself to be voting against is exactly what my opponent takes
himself to be voting in favor of. Otherwise, the idea that our votes, on
a given occasion, are to be aggregated and weighed against one an-
other becomes a nonsense.
What I have just described is rudimentary by comparison with the
processes employed in actual legislative assemblies such as the Con-
gress of the United States. Bills are longer and more complex than
the sort of motions one hears at faculty meetings. They have usually
been drafted-more or less competently-in advance, and there are
many stages of deliberation (including committee stages, whose pro-
ceedings may be much less formal) that bills must go through before
they are adopted. And, this is to say nothing of the vicissitudes of
bicamerality, conference committees, and the rest.
For the most part, however, these complications enhance the

need for a determinate text to focus and coordinate the various stages
of the legislative process. Without a text to consider, to mark up, to
amend, to confer about, and to vote upon, the process of law-making
in a large and unwieldy assembly would have even a greater air of
babel-like futility than that which is currently associated with
Congress.
Thus, whether we are talking about a small-scale meeting or a
large-scale legislative process, the positing of a formulated text as the
resolution under discussion provides a focus for the ordering of delib-
eration at every stage. The existence of a verbalized bill, motion, or
resolution is key to norms of relevance, and key to the sense, which
procedural rules are supposed to provide, that participants' contribu-
tions are relevant to one another and that they are not talking at cross
purposes. Maybe, a one-person deliberative body can do without
this-though even there, many of us are familiar with the mnemonic
virtues of a formulated proposition in our own solitary decision-mak-
ing. And maybe, decision-making in a small group of oligarchs or in a
junta of familiars can do without this as well, if they can move toward
consensus on the basis of conversational informality. But the sense of
a determinate focus for discussion-something whose existence is dis-
tinct from the will or tacit understandings of particular members'-
seems absolutely indispensable for a large and diverse assembly of
people whose knowledge and trust of one another is limited.
VIII
If there is anything to this hypothesis, then we might want to start
thinking about the textual canonicity of legislation in a slightly differ-
ent way. I said in Part I that one of the values most commonly associ-
ated in the modern world with legislation is democratic legitimacy:
We should defer to statutes because they have been enacted by a dem-
ocratically elected entity. Just as the idea of democracy is insufficient
to explain why we prefer a large elected legislature to a single elected
legislator, so the democratic principle is insufficient to explain the
particular way in which authority is accorded to legislation in the mod-
88. Compare the characterization of the advantages of a written constitution in

HANNAH ARENDT, ON REVOLUTION 156 (1965):
[T]he Constitution [is] a written document, an endurable objective thing, which,
to be sure, one could approach from many different angles and upon which one
could impose many different interpretations, which one could change and
amend in accordance with circumstances, but which nevertheless was never a sub-
jective state of mind, like the will.
MARYLAND LAW REVIEW [VOL. 54:633
ern world, viz., by taking seriously the exact words that were used in
the formulations that emerged from the legislative chamber. If I am
right, we now have an explanation for the importance of the ipsissima
verba which is oriented primarily to the legislators' dealings among
themselves, rather than directly to the issue of their collective author-
ity vis-a-vis the people.
The final step, then, in pursuit of this hypothesis would be to
show how this account of the importance of a text to the legislators is
connected with the authority of the text for its intended audience.
Here there are a couple of lines to pursue. First, as we have seen, the
existence of orderly discussion is necessary to secure whatever Aristo-
telian advantages accrue from deliberation in a large and diverse
group. Unless the diverse experiences and knowledge of the various
legislators can connect and be synthesized, it is unlikely that their in-
teraction will produce standards that are superior to those that any
individual citizen could work out for herself. The conditions for or-
derly discussion, then, are indirectly conditions for the legislature's
authority, in the Razian sense.8 9 In other words, authority requires
superior expertise; superior expertise comes from deliberation among
those who are different from one another; deliberation among those
who are different from one another is possible only on the basis of
formal rules of order; and crucial to rules of order is the postulation
of an agreed text as the focus of discussion.
Second, respect for statute law is partly a matter of respect for the
legislature as a forum whose representativeness is an aspect of the fair-
ness 90 of the way a community makes its decisions. To the extent that
representativeness requires diversity in the assembly, respect for that
fairness is a matter of respecting the conditions under which diverse
representatives can deliberate coherently. Thus, fairness-based re-
spect for the legislature as a body may require not only that we respect
the standards which it posits, but also that we respect these more for-
mal aspects of the way in which its posited standards are arrived at-
89. See RAZ, supra note 44. Professor Raz sets out the following thesis about the justifi-
cation of authority claims:
[T]he normal way to establish that a person has authority over another person
involves showing that the alleged subject is likely better to comply with reasons
which apply to him . . . if he accepts the directives of the alleged authority as
authoritatively binding and tries to follow them, rather than trying to follow the
reasons which apply to him directly.
Id. at 53.
90. Fairness, in the sense defined by DwoRmw, LAW's EMPIRE, supra note 41, at 164-65.
and thus that we respect the standards in question under the auspices
of text-based formality.9 '
Ix
It is unlikely that the account just given is the full story or the only
explanation of the peculiarly formal or text-based way in which we
approach pieces of legislation. But it furnishes a glimpse of the op-
portunities that may open up for jurisprudence when we pose ques-
tions about the composition of legislative institutions and take
seriously the possibility that law made by an assembly of persons may
differ significantly, in concept as well as in quality, from law made by
one person.
The argument that I have developed also indicates the possibility
of closer ties between jurisprudence and democratic theory than have
existed hitherto. Too often, analytical jurisprudence in the common
law tradition, with its emphasis on courts and its depiction of the legis-
lature as a disreputable, embarrassing and inconvenient sideshow,
contrives to present the political provenance of our laws as a matter
unworthy of philosophical notice. The impression we are given is that
the question of provenance goes to the political credentials of the leg-
islator, which may in turn determine the substantive content of legisla-
tion; but provenance appears largely irrelevant to the status of
legislation as law or to the conception of law-as, for example, a gen-
eral rule, a moral principle, or an interpretable text-which legisla-
tion exhibits. In the last analysis, that impression may be correct; that
is, it may be the case that we can do all the philosophical and concep-
tual thinking that we need to do about law while ignoring the issue of
who makes law and on what authority. But it is not something we are
entitled to assume a priori. It is surely worth exploring the possibility
that the nature of law reflects the conditions under which it is made.
A legal system which vests final authority in an assembly comprising
hundreds of representatives-and which does so as a way of expres-
sing principles of popular sovereignty, self-government and demo-
cratic self-determination-may require a somewhat different sort of
jurisprudence from that appropriate to a system dominated by the
edicts a single, rational law-giver.
91. A fuller account of this respect would present it not only as a theory of general civic
obligation to the legislature, but also as an issue of the respect owed to the legislature
specifically by the courts, in the context of democratic legitimacy and the separation of
powers.

The Dignity of Legislation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Dignity of Legislation

Uploaded by

Copyright:

Available Formats

Maryland Law Review

Volume 54 | Issue 2 Article 12

The Dignity of Legislation

Follow this and additional works at: http://digitalcommons.law.umaryland.edu/mlr

Civil Society being a State of Peace, amongst those who are

Locke as the great proponent of natural right and accordingly as a

Locke's hypothesis is a challenging one, but it is not the chal-

4. LocKE, supra note 1, at 373 (Second Treatise 134).

provides a sort of infrastructure or underpinning -for Locke's idea,

9. Cf the definition of "legislatures" in THE BLACKwELL ENCYCLOPEDIA OF POLITICAL

sembly ought to be a sort of microcosm, mapping to scale the features

I labor this point, to emphasize that the constitutional preference

assembly, the more

15. CONDORCET: SELECTED WITINGs 49 (Keith M. Baker ed., Bobbs-Merrill 1976)

does anything;' and yet we are governed by the House of Commons-

25. Id. at 13940.

ticularly from positivist jurisprudence, which is reputed to regard de-

and substantive norms in the course of qualifying that person or body

36. HART, supra note 32, at 69, 144-48.

whose composition reflects electoral competition, each elected mem-

not necessarily be directed at legislators themselves, telling them how

In this regard, our jurisprudence is badly handicapped by a the-

ical philosophy, the sovereign was almost always referred to as "he."47

though it had the character of a single author of the texts we call

the many acting in a representative assembly. We want our philoso-

54. See HART, supra note 32, at 54-56, 83-88.

eenth and nineteenth centuries regarded legislation-by an enlight-

In much the same way, I think we should be open to the possibil-

problem of interpreting and applying a statute is usually presented as

To say this is not to deny that statutes need interpretation or that

of institution a legislature is: a large gathering of disparate individuals

70. PrTKIN, supra note 10, and accompanying text.

quite misleading so far as our models of political deliberation are

enhanced by the very features of their situation that make it impor-

ically to the challenges and difficulties of that diversity.8 3 Analytic ju-

In circumstances of this kind, we see the importance of delibera-

My hunch is that textual canonicity and the procedural formality

87. HOBBES, supra note 48, at 190.

No doubt, in the course of discussion, someone may feel that it

For the most part, however, these complications enhance the

88. Compare the characterization of the advantages of a written constitution in

You might also like